|
|
||||||||
Ann Thorac Surg 2007;83:153-160
© 2007 The Society of Thoracic Surgeons
First Department of Surgery, Hirosaki University School of Medicine, Aomori, Japan
Accepted for publication August 8, 2006.
* Address correspondence to Dr Fukuda, First Department of Surgery, Hirosaki University School of Medicine, 5 Zaifu-cho, Hirosaki, Aomori 36-8562 Japan. (Email: ikuofuku{at}cc.hirosaki-u.ac.jp).
| Abstract |
|---|
|
|
|---|
METHODS: Rabbits were assigned four groups: CPB for 60 minutes, control CPB group; low-dose lipopolysaccharide (LPS) administration without CPB, LPS group; CPB after lipopolysaccharide administration, LPS+CPB group; or preparation with continuous infusion of sivelestat and CPBs after lipopolysaccharide administration, EI group. Blood samples to determine blood gas concentration, plasma elastase activity, and plasma interleukin-8 levels were obtained. Histopathologic examinations of the lung were performed.
RESULTS: The decreased arterial oxygen pressure at the end of CPB was observed in the LPS+CPB group only, but was suppressed in the EI group (p < 0.01). Elastase activity was markedly elevated at 120 minutes after CPB, and interleukin-8 levels were markedly elevated at 180 minutes in the LPS+CPB group but were much lower (p < 0.05) in the EI group. Histopathology demonstrated accumulation of polymorphonuclear neutrophils in bronchoalveolar areas in the LPS+CPB group (p < 0.01). Pulmonary myeloperoxidase activity was significantly lower in the LPS+CPB group than in the other groups (p < 0.01). These changes were minimal in the EI group.
CONCLUSIONS: The combination of low dose LPS+60 minutes of CPB, but neither intervention alone, produced evidence of acute lung injury in a rabbit model. This did not occur when the animals were pretreated with sivelestat.
| Introduction |
|---|
|
|
|---|
Recent progress in biocompatible CPB circuits made the incidence of this clinical syndrome low. However, severe respiratory failure occurs frequently in patients with shock or infection. In these patients, one of prophylactic strategies is pharmacologic intervention to suppress inflammatory reaction.
Sivelestat sodium hydrate is a specific inhibitor of neutrophil elastase with a small molecular weight [2]. It inhibits neutrophil elastase activity competitively but does not affect other serine proteases released by PMNs. The effects of on endotoxin-induced lung injury, postperfusion lung, and ischemiareperfusion have been investigated in several studies [35]. In a recent study, elastase inhibition with sivelestat reduced inflammatory mediators and preserved neutrophil deformability during simulated extracorporeal circulation [6]. We hypothesized that the administration of sivelestat before CPB can prevent the progression to acute lung injury after CPB. To test this hypothesis, we established a rabbit model for acute lung injury after CPB and evaluated the effect of sivelestat.
| Material and Methods |
|---|
|
|
|---|
Experimental Groups
Twenty Japanese white rabbits weighing 3 to 4 kg were used in the study. Five rabbits each were randomly allocated to one of four groups (Fig 1). In the control CPB group, rabbits were heparinized and placed on CPB for 60 minutes, then taken off CPB, given protamine, and monitored for 180 minutes. In the low-dose lipopolysaccharide (LPS) group, a low dose of LPS (0.5 µg/kg) was infused. Rabbits underwent median sternotomy for 60 minutes, and were monitored for 180 min. These animals received no heparin or protamine. Rabbits in the LPS+CPB group were infused with low-dose LPS, heparinized, and placed on CPB for 60 minutes, taken off CPB, given protamine, and were monitored for 180 minutes. In the EI group, rabbits were infused with sivelestat (10 mg/[kg · h]) continuously, and the protocol of LPS and CPB were same as the LPS+CPB group. Animals were heparinized and placed on CPB for 60 minutes, taken off CPB, given protamine, and monitored for 180 minutes.
|
Rabbit Cardiopulmonary Bypass Model
After anesthesia induction with intravenous ketamine (10 mg/kg), rabbits were placed in the supine position and a 4.0-mm endotracheal tube was connected to the ventilator through a tracheostomy. Anesthesia was maintained with continuous infusion of pentobarbiturate (5 mg/[kg · h]) and bolus infusions of vecuronium bromide (0.5 mg/[kg · h]) and ketamine (5 mg/[kg · h]). Rabbits were ventilated using a pressure control ventilator. In all protocols, inspiratory time was 0.8 seconds, peak inspiratory pressure was 12 cm H2O, and positive end-expiratory pressure was 1 cm H2O. Before and after CPB, fraction of inspired oxygen (FIO
2) was 0.5, and the respiratory rate was adjusted from 30 to 40 to maintain normocapnia. During CPB, FIO
2 was 0.2 and the respiratory rate was maintained at 10 breaths/minute.
The left jugular vein and the right femoral artery were cannulated for obtaining blood samples, monitoring mean arterial pressure, and blood gas analysis. The extracorporeal circuit consisted of polyvinylchloride tubing, a venous reservoir, a membrane oxygenator (MERA, Inc, Tokyo, Japan), and a roller head pump.
After median sternotomy and administration of heparin (300 U/kg), an arterial cannula was introduced into the right brachiocephalic artery and a venous cannula was introduced into the right atrium via the right atrial appendage. Normothermic CPB was initiated at a flow rate of 80 to 100 mL/(kg · min) and maintained for 60 minutes. Mean arterial pressure was maintained at 50 to 60 mm Hg by adjusting the flow rate. Systemic heparinization was monitored by activated coagulation time and it was maintained at a minimum of 400 seconds.
When hypotension occurred in weaning from CPB, infusions of extracellular fluid type were used and catecholamines (dopamine or epinephrine, or both) were given to maintain an adequate blood pressure or heart rate. We used a warming blanket under the rabbits to maintain the temperature at about 37°C. Protamine (3 mg/kg) was given after removal of the cannulas. Blood in the extracorporeal circuit was transfused over a period of 30 minute without a homologous blood transfusion. Rabbits were observed for 180 minutes after CPB termination.
Blood samples were obtained at anesthesia induction (baseline), at 2 minutes after heparinization, at 60 minutes after the start of CPB, and at 30, 60, 120, and 180 minutes after CPB termination. At the end of the experiment, the rabbits were euthanized by an overdose of pentobarbiturate and the left lung was removed immediately.
Endotoxin Infusion
Low-dose LPS (0.5 µg/kg) was administered to 15 rabbits through a marginal ear vein during a period of 20 minutes before median sternotomy in the three experimental groups, but not in the control CPB group.
Sivelestat Administration
An infusion pump was used to deliver a continuous intravenous administration of sivelestat (10 mg/[kg · h]) in 5 rabbits from induction of anesthesia to the end of the experiments.
Measurements
Blood counts and blood gases were measured using fully automated blood cell and blood gas analyzers, respectively. Plasma was separated by centrifugation at 2000g for 10 minutes at 4°C after being cooled on ice for 15 minutes and was stored at80°C until analysis. Plasma elastase activity was determined with the synthetic substrate N-methoxysuccinyl-Ala-Ala-Pro-Val p-nitroaniline, which is highly specific for neutrophil elastase and is not hydrolyzed by cathepsin G [7].
Plasma interleukin-8 (IL-8) was measured using sandwich enzyme-linked immuno sorbent assay (ELISA) kits according to the manufacturers instructions (BioSource International, Inc, Camarillo, Calif).
Pulmonary myeloperoxidase activity in the peripheral part of the left lower lobe was assayed to assess the number of recruited neutrophils to the lungs in each group. After sacrifice, a part of the left lower lobe lung was frozen and stored at80°C until analysis. Samples were homogenized and the supernatant fluids were measured with a SUMILON peroxidase assay kit (Sumitomo Bakelite, Tokyo, Japan) [8]. Myeloperoxidase activity per gram of pulmonary tissue was expressed as units per gram.
Histopathologic Examinations
Ten percent formalin-fixed, paraffin-embedded lung tissue sections were prepared. From these specimens, four slides for each lung were prepared. Two were stained with hematoxylin and eosin and examined under a light microscope. The remaining two were stained with naphthol AS-D chloroacetate esterase that stains granules in granulocytes as red. The pulmonary PMN count in proximal hilar (bronchoalveolar) areas where small bronchioles and pulmonary arterioles were observed microscopically and in peripheral (alveolar) areas was quantified in 20 random fields (x400) per slide in a blinded manner.
Statistical Analysis
All data are expressed as mean ± standard error of the mean. Statistical analysis was performed using SPSS (SPSS, Inc, Chicago, Ill) for Windows XP (Microsoft Corp, Redmond, Wash). Repeated measures two-way analysis of variance (ANOVA), followed by the Bonferroni multiple comparison, was used to analyze group and time effects in blood sample data. When group effects were significant (p < 0.05), two-way ANOVA between control and each experimental group (separately) was used to establish significant differences (p < 0.05). The unpaired t statistic was used for specific comparisons at specific time points between groups when the two-way ANOVA group (with Bonferroni adjustment) effect was significant. One-way ANOVA, followed by the Bonferroni test, was used to analyze group effects in the histologic pulmonary PMN count and pulmonary myeloperoxidase activity. Differences were considered statistically significant at p < 0.05.
| Results |
|---|
|
|
|---|
|
|
Plasma elastase activity did not change in the control CPB, LPS, and EI groups during the entire course. Elastase activity in the LPS+CPB group gradually increased after CPB termination, however, and was significantly higher than in the other groups at 120 and 180 minutes after CPB termination (Table 1, Fig 2B).
The plasma IL-8 level did not change in the control CPB and LPS groups, but progressively increased in the LPS+CPB group and was significantly higher than in the other groups at 60, 120, and 180 minutes after CPB termination. Although the IL-8 level in the EI group increased temporarily at 30 minutes after CPB termination and was significantly higher than in the control CPB and LPS groups, it then decreased gradually (Table 1, Fig 2C).
Histopathologic examination showed that the pulmonary PMN count in bronchoalveolar areas was significantly higher in the LPS+CPB group than in the other groups. In the EI group, this value was significantly higher than in the control CPB and LPS groups. The PMN count in the alveolar areas was lowest in the LPS+CPB group (Fig 3A). There were no significant difference between the control CPB and the LPS+CPB groups, but the LPS+CPB group was lower than the EI group and LPS group (p < 0.05). Pulmonary myeloperoxidase activity was significantly lower in the LPS+CPB group than in the other groups (Fig 3B).
|
|
| Comment |
|---|
|
|
|---|
Acute lung injury after cardiac surgery was shown to develop in 0.4% to 3% of CPB patients, but mortality remained high, at 15% to 70% [912]. The severity of postoperative SIRS and the possibilities of developing acute lung injury or multiple organ failure, or both, may be related to precedent or subsequent insults such as endotoxemia, hypoxia, and ischemiareperfusion injury that would not cause clinically significant results individually. Among those insults, endotoxemia is especially important, because it is the strongest trigger for the development of acute lung injury and its mortality and morbidity remain very high [13]. Most patients are exposed to endotoxin during cardiac operations. The origin of exposure to endotoxins during surgical procedures remains unconfirmed, but current theories suggest that high levels of circulating endotoxins during CPB are caused by gut hypoperfusion, with resulting translocation of bacteria from intestinal mucosa [14].
We established a new endotoxin-induced acute lung injury model after CPB in the rabbit. Many studies on an endotoxin-induced lung injury model in rabbits used LPS, and the amount of LPS used was from 100 µg/kg to 5 mg/kg. But in situations where there is a combination of endotoxin and another insult, the required amount of LPS to induce lung injury is extremely small. The combination of low-dose LPS (15 to 20 µg/kg) and platelet-activating factor caused acute lung injury in rabbits [15], and the combination of very low-dose LPS (1µg/kg) and CPB caused acute lung injury in the porcine models [16].
In our study, the control CPB and LPS groups did not show a decrease in PaO 2/FIO 2 or other significant morbidity. However, the LPS+CPB group showed a significant decrease of PaO 2/FIO 2 from CPB termination to the end of the experiment. In addition, the hemodynamic state deteriorated significantly after CPB termination. This model demonstrated that a very small amount of endotoxin can cause development of acute lung injury after CPB use.
Activated PMNs, monocytes, and macrophages produce IL-8; and neutrophil elastase released from PMNs induces IL-8 expression [17]. An elevation of plasma IL-8 level may be related to cardiac and pulmonary dysfunction after CPB. We therefore measured the plasma IL-8 level as an important marker to evaluate the effect of sivelestat on acute lung injury after CPB. Elastase activity was significantly higher in the LPS+CPB group after CPB than in the other groups. Activation of monocytes stimulated from LSP/LPS-biding-protein complex occurs immediately by gene expression of cytokines through the extracellular signal-regulated kinases pathway [18]. In a simplified contact activation model by Matsuzaki and colleagues [6], release of neutrophil F-actin and IL-8 occurred in 30 minutes and 90 minutes after start of simulated extracorporeal circulation, respectively. When PMNs are nonspecifically activated by cytokines and ready for release of neutrophil elastase, the reaction would occur within shorter time.
Neutrophil elastase itself cleaves the phosphatidylserine receptor on macrophages, which will clear off activated but apoptotic PMNs [19]. It results in dissemination of neutrophil elastase due to autolysis of apoptotic PMNs, which should otherwise be digested by macrophages. The failure of phagocytosis for activated PMNs will make an explosive malignant circle, as shown in this model. Because platelet counts were extremely decreased 30 minutes after CPB, we suppose the LPS+CPB group finally fell into the state of disseminated intravascular coagulation after 60 minutes of CPB. These results suggest that sivelestat can reduce the enhanced inflammatory response by interrupting the chain of reaction between IL-8 and neutrophil elastase and that sivelestat can inhibit the vicious cycle promoted by LPS and CPB.
Direct injury to the lungs was evaluated by pulmonary myeloperoxidase activity and histopathologic examination. Myeloperoxidase is released from PMNs and indicates production of reactive oxygen species and hypochlorous acid. In the LPS+CPB group, PMNs were sequestered in vessels as aggregation and accumulated in interstitial spaces in proximal (bronchoalveolar) areas. In contrast, little accumulation was observed in peripheral (alveolar) areas, and this finding was related to the low myeloperoxidase activity in the LPS+CPB group.
The continuous administration of sivelestat inhibited PMN aggregation of proximal vessels and PMN accumulation, and it attenuated the respiratory and hemodynamic devastation observed in the LPS+CPB group. In addition to platelet and coagulation system abnormalities directly caused by CPB, further endothelial dysfunction via several cascades, including activated PMNs, occurs after CPB with endotoxemia. PMN activity is enhanced rapidly so that aggregations of circulating PMNs are formed and occlude the pulmonary capillaries.
For the prevention or treatment of SIRS and acute lung injury after CPB, several distinct strategies were used to protect against injury related to PMNs. Leukocyte depletion using a leukocyte filter or circuits with copolymer surfaces contributed to some improvements in experimental CPB and clinical studies [2022]. Although they may be useful, it is probably difficult to improve the clinical course if other insults are added to CPB. This is because a large percentage of PMNs have already strongly adhered to the endothelium, have accumulated in pulmonary tissue, and have released elastase by the time a diagnosis of acute lung injury is made.
Sivelestat inhibits neutrophil elastase activity competitively. An intrinsic antiprotease with a large molecular weight that inhibits neutrophil elastase in an ordinary status is inactivated immediately by reactive oxygen species in an inflammatory status. Sivelestat is not inactivated by reactive oxygen species, and advantages of sivelestat compared with neutrophil depletion and steroids are that it can enter the microenvironmental spaces between PMNs and their substrate tissues owing to structural constraints and specifically inhibit tissue-bound neutrophil elastase, the terminal production in cytotoxic reaction [23]. Sivelestat may reduce the vicious cycle leading to the development of SIRS and acute lung injury. Although some studies in the clinical use of sivelestat for acute lung injury demonstrated that administration of sivelestat after acute lung injury was significant, it had no effect on 28-day all-cause mortality or the duration of mechanical ventilation [24]. Administration of sivelestat before significant acute lung injury is expected to improve the clinical course.
In conclusion, the present study showed that the administration of sivelestat before CPB can prevent LPS induced acute lung injury through improvements of oxygenation, amelioration of platelet depletion, and reduction of PMN accumulation/sequestration in the rabbit model. This strategy of administration may be useful in patients at risk for postoperative acute lung injury undergoing cardiac surgery.
| Acknowledgments |
|---|
|
|
|---|
| References |
|---|
|
|
|---|
Related Article
This article has been cited by other articles:
![]() |
S. Iwamoto, A. Higashi, T. Ueno, M. Goto, Y. Iguro, and R. Sakata Protective effect of sivelestat sodium hydrate (ONO-5046) on ischemic spinal cord injury Interactive CardioVascular and Thoracic Surgery, June 1, 2009; 8(6): 606 - 609. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Ando, T. Murai, and Y. Takahashi The effect of sivelestat sodium on post-cardiopulmonary bypass acute lung injury in a neonatal piglet model Interactive CardioVascular and Thoracic Surgery, October 1, 2008; 7(5): 785 - 788. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Alipour, A. Omri, M. G. Smith, and Z. E. Suntres Prophylactic effect of liposomal N-acetylcysteine against LPS-induced liver injuries Innate Immunity, October 1, 2007; 13(5): 297 - 304. [Abstract] [PDF] |
||||
![]() |
M. E. Jessen Invited commentary Ann. Thorac. Surg., January 1, 2007; 83(1): 160 - 160. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |